JASN_DE

Which isn’t a great idea with all the breaking changes. I’d assume it gets better after v2, but still.

Write error : cannot write block : No space left on device

Seems like your boot partition is full.

It should, but apparently it isn’t, at least not reliably.

ZFS

Goaccess reads the Traefik log and spits out an HTML file you can then serve via e.g. Nginx. Works for me.

Realistically though, you could still use a VPN and have it be pretty easy for your family members IF you have access to their router console and IF said router supports network wide wireguard or openVPN connections. Having both networks tied in to eachother that way makes it so that nobody ever has to use a VPN client to connect, but still only devices from their network (or yours) will be able to connect.

Realistically this plan dies the moment someone takes their phone outside of the WiFi range. It’s fine in theory, but fails miserably in non-techie real life.

There is no need to run Traefik if you already have a working reverse proxy. Also, unless your nginx is running on non-standard ports, the ports 80 and 443 should not be available for traefik, which will prevent it from working correctly.

Most of them are simply irrelevant to me. Yes, the book might be award-worthy, but I can virtually guarantee that I have differing preferences than the committees. The more “highbrow” the further I’m usually from their rating.

You can narrow it down: attach to the container, this’ll give you a life “feed” of all messages the container produces. Then with that running, open your app and see if the container has anything to say about the sync process.

As always: have you checked the logs?

Not for a Synology, but I doubt Hetzner cares where the data comes from. Works well, especially once you got the keys in place. Have stored and restored, it’s a simple file storage.

Do you think the fediverse has a civility problem

This has little to do with the Fediverse itself and mostly with the people involved. It would be the same on any other platform, so simply switching the software stack or whatever would do exactly zilch.

A fraction of people are going to be assholes. Sad but inevitable.

and is there anything we can do about it?

You can ban the dickheads, but that’ll only get you so far.

As much as I hate it, I think I’ll have to go through the logs after all.

Selfhosting without checking the logs to hunt down errors? That’s not going to be pretty.

Let’s Encrypt is fully automated and will issue certificates as long as you provide an email address AND have a proper, working config. Don’t get stuck on that email “issue”, your problems will lie somewhere else.

As always when problems arise: check the log files.

I haven’t tested that part of it yet, but the self-hostable StirlingPDF offers conversion from PDF to a number of formats.

The rest I use it for works fine, so maybe that could be an option.

If you’re doing a fresh instance it will solve a lot of issues. Personally I run a Nextcloud instance which got its own 2TB SSD. I mounted the disk at /nextcloud, then used bind mounts in docker compose for db and NC.

Which part is your problem, serving the media from disk, or transcoding and serving that stream?

A big portion of that is caused by the drives, so you’d have to compare the empty QNAP vs your empty machine. Also, depending on which NAS appliance, check that the CPU is actually powerful enough to run all your services.

will all my Jellyfin traffic go through the VPS and count as bandwidth used?

Yes.

“Cloud” simply means it’s on other people’s machines.